Recently I’ve been learning about NodeJS and how to write server side JavaScript. This week I was creating a route that pulled variables from POST, inserted them into a SQL database, and then retrieved the ID of the inserted record. Later on I wanted to take that ID and retrieve the record submitted. When the code ran I didn’t have errors but I had symptoms that didn’t make sense. As my debugging strategy I started logging variables to make sure they had values that I expected.
sqlInsertServer = "INSERT INTO serverbuilddata (ServerName, CPUCount, RAMCount) VALUES ('" + serverName + "'," + CPUCount + "," + serverMemory + ');'
console.log('SQL Statement: ' + sqlInsertServer);
var buildID;
con.query(sqlInsertServer, function (err, result) {
if (err) throw(err, null);
buildID = result.insertId;
console.log("Inside function. Server record inserted. Build ID: " + buildID);
);
console.log(buildID);
The first instance of build ID I logged inside the function returned the correct result, but the one outside of the function was undefined. I thought it was a scoping problem at first and tried to make sure the variable was declared globally. Then I noticed that the undefined variable outside the function was being logged before the one in the function. Then it kind of clicked about how Node is asynchronous. It doesn’t wait for other functions to run before the code gets executed even if they are in the correct order. After some research about my options I decided to go with call backs since they seemed to be the most recommended and best way.
sqlInsertServer = "INSERT INTO serverbuilddata (ServerName, CPUCount, RAMCount) VALUES ('" + serverName + "'," + CPUCount + "," + serverMemory + ');'
console.log('SQL Statement: ' + sqlInsertServer);
function serverInsert(callback) {
var buildID;
con.query(sqlInsertServer, function (err, result) {
if (err) callback(err, null);
buildID = result.insertId;
callback(null, buildID);
console.log("Inside function. Server record inserted. Build ID: " + buildID);
});
}
serverInsert(function (err, buildID) {
if (err) {
// error handling code goes here
console.log("ERROR : " + err);
} else {
// code to execute on data retrieval
console.log("result from callback is : " + buildID);
})
Callbacks wait for the function to provide it with a value. The problem is that value still can’t be used outside the function. And if that value in the callback needs to be used in another SQL function you need to nest another callback in the callback. Before you know it you have a mess. It does work but I’m still looking for better ways to do it. I’ve read about promises but I’m not ready for that yet.
